SDx

SDx (Software-defined everything) is a term used to describe how virtualization and abstracting workloads from the underlying hardware can be used to make information technology (IT) infrastructures more flexible and agile. It is also a term used to describe the trend towards technologies such as software-defined networking, software-defined storage, and software-defined data centers. As well, it describes the shift in physical items and functions that are becoming software-defined, and the implications of this evolution.

Some see SDx as having a mission to break down discrete IT silos, such as compute, storage, and networking, by using software to bridge technological and organizational gaps. However, what SDx has shown is that it is capable of delivering costs savings and greater speed of delivery compared to hardware implementations. Others link the emergence of SDx approach to the Internet of Things (IoT), which supports the basic definition of SDx as a future system where one set of software rules over numerous connected machines while directing many types of user activity. This approach has also concerned some, as expanding network capability can also generate greater security gaps.

The use of SDx has been shown to have an impact on cloud, infrastructure, and automation, in different ways and processes.

Visualization of the umbrella of Software defined everything (SDx)

In infrastructure technology, there have been many innovations and developments that have already evolved IT infrastructure, which includes storage, networking, security, and compute, towards macro needs, and have seen a move towards abstraction over commodity hardware. In order to meet the needs and expectations of consumers and end-users, SDx applications need flexibility, personalization, security, and efficiency to scale, which means they also need the proper infrastructure support to do it. Traditional, hardware-centric infrastructures make meeting these requirements difficult if not altogether impossible. This has led to the implementation of IT infrastructure as software with the support of standard hardware. These systems include the following:

In Software defined network (SDN), the network architecture is abstracted to make network devices more programmable. This abstraction can be used to allow an administrator to quickly respond to changing business requirements.

In software defined storage (SDS), the physical hardware is decoupled and uses storage media, such as NVMe, SSD, or HDDs, in server hardware to provide virtual storage. This decoupling can provide an administrator to pool and manage storage resources through policies and administered configurations, rather than configuring single drives for different events and uses.

The virtualization of servers allow users to save hardware and software costs and provide virtual compute resources to multiple users, allowing for greater flexibility in resource utilization. This can also be used to deploy and configure individual compute environments, and to automate as much background compute work as possible.

Software defined data centers (SDDC) offers an analytics-driven approach to balancing the resources that programs require in virtual and cloud computing environments, which is also called composable infrastructure.

With staff and employees moving increasingly online and mobile, software defined access solutions offer end-to-end traffic segmentation to both streamline authorization and increase total security and to allow employees to use mobile devices to request data access from within the cloud or on the edge of an IT stack.

Organizations continue to look for new ways to deliver branch-level benefits at scale and on demand. Software defined branch solutions have made it possible to consolidate hardware network functions into a single, streamlined software platform.

The range of demand for IT services and computer system resources are delivered, through cloud architecture, to organizations or customers per their requirements over the internet. Cloud architectures typically rely on software-centric approaches because the solutions favor self-service and dynamic rapid deployments. Cloud architectures tend to use similar design templates and reusable templates for users to outsource non-essential elements. And, unlike traditional IT infrastructure, cloud architectures can scale horizontally across servers and racks, and scale up and down in terms of CPU or resource consumption while being capable of tolerating component failure.

The range of services delivered through the cloud, and through software abstraction, include infrastructures as a service, platform as a service, software as a service, function as a service, and backup as a service. Each of these services are delivered over the internet and allow users to take advantage of the flexible environment which automates many portions of the load balancing and scaling, to make them easier for the end-users.

Automation refers to a code approach that programs different functions to work automatically, and enables IT to be more flexible, fast, secure, and stable. This automation can be used in infrastructure to define and manage IT infrastructure with configurable software. It can be used to built and test to automate these processes and to keep the results of built and test traceable. It can be used in installation to allow the configuration of operating systems and the installation of applications to be automated. And software defined automation can be used to allow automatic monitoring in a data collection process, for use in identifying or correcting errors, for example.

Proprietary hardware is often though to create systemic solution silos, making what works for a given department not always capable of working with the hardware used in other departments, which can create many difficulties for IT teams, requiring duplication in time and effort to manage multiple technology stacks. SDx is considered to be capable of solving these problems by offering three key advantages:

• Agnostic implementation—traditional IT deployments have previously relied on consistency at the hardware level to create a corporate network. Virtualizing these operational layers, SDx allows these environments to expand without concern over interoperability, through an agnostic implementation, and common hardware capabilities are used to deliver a consistent experience.
• Automated operation—the virtualization of key processes can offer standardization that can increase the possibilities for automation. The standardization allows enterprises to streamline operations such as resource assignment or network traffic management, both reducing the potential for human error and allowing IT teams to focus their efforts to more important areas.
• Adaptable application—IT teams can customize SDx applications across physical and virtual distance by decoupling resource management from underlying hardware, and allowing departments to retain unique functionality without increasing complexity.

With the use of software defined everything, there are security concerns, as software has been shown to be hackable and porous in the past. For example, in the case of a software defined network (SDN), while it provides a virtualized network, and looks more like the future norm of networking, if the SDN controller or application are compromised, an attacker can hack the SDN applications and manipulate security applications to reprogram the network traffic flow through the SDN controller.

And as server virtualization, mobility, and cloud computing have become new norms for changing business needs, the traditional architecture has begun to fall short to meet the network demands, pushing towards increased use of SDN to meet those demands. Software defined security offers a solution to these concerns, and is a model in which information security is controlled. The functions of network security devices, such as firewalling, intrusion detection, access controls, and network segmentation are extracted from the hardware devices to be virtualized in a software layer, as other processes in SDx. And the virtualization of these functions can offer additional functionality in order to add new layers of security.